The main goal of this project is to develop sensitive neurobiological markers of incipient and mild Alzheimer's disease (AD). With quantitative structural, magnetic imaging (fMRI) techniques, it is now possible to detect and quantify, in vivo, the pattern of anatomical pathology during different stages of degenerative diseases such as AD. In behavioral symptoms in AD and b) examine the specific role of certain brain structures in human memory function because of the age- or disease-related occurrence of "lesions" in these structures. The first major aim of the project is to determine longitudinally, in each individual, with serial MRI scans and quantitative volumetric analysis, the earliest sites of anatomical pathology in AD and the pattern of spread of such pathology. Because of the opportunities that a longitudinal design affords, we will also determine whether the sequence of anatomical changes observed correspond to the sequence of alterations in behaviors critically dependent on those neural evolution of symptoms (and their heterogeneity) in the disease. An important novel feature of the proposal is the use of diffusion tensor imaging (DTI) to define and quantify deterioration of white matter as a function of age and disease progression. This is a new structural imaging technique that can detect alterations in the microstructure of white matter. Research on the anatomical and functional progression of AD is important for an understanding of the pathophysiology of the disease. The development of sensitive early biological markers that differentiate those at risk for developing AD is critical, since therapeutic strategies may be more effective in the incipient phase of the disease.